when I am passing manually json file as key to load in python shell then its working fine. code below
import os
import json
import boto3
import io
import requests
import botocore
bucket_name = 'dev-data'
folder_name = 'raw/test/'
key_source = 'raw/test/extract_api_20200719.json'
s3_client = boto3.client('s3')
json_obj = s3_client.get_object(Bucket=bucket_name, Key=key_source)
json_data = json_obj["Body"].read().decode('utf-8')
print("############################json_data####################### :", json_data )
print("############################json_data_type################## :", type(json_data))
json_dict = json.loads(json_data)
print("############################json_dict####################### :", json_dict )
print("############################json_dict_type ################# :", type(json_dict))
However when using for loop to read JSON object from s3 bucket, then I am getting the error
import os
import json
import boto3
import io
import requests
import botocore
bucket_name = 'dev-data'
folder_name = 'raw/test/'
s3_resource = boto3.resource('s3')
bucket = s3_resource.Bucket(bucket_name)
for obj in bucket.objects.filter(Prefix=folder_name):
print('Object to extract :', obj)
print('obj key: ', obj.key)
s3_client = boto3.client('s3')
json_obj = s3_client.get_object(Bucket=bucket_name, Key=obj.key)
json_data = json_obj["Body"].read().decode('utf-8')
json_dict = json.loads(json_data)
error:
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)
Advertisement
Answer
Some of the entries in bucket.objects don’t have any JSON in the data, so check for that and skip them.
for obj in bucket.objects.filter(Prefix=folder_name):
print('Object to extract :', obj)
print('obj key: ', obj.key)
s3_client = boto3.client('s3')
json_obj = s3_client.get_object(Bucket=bucket_name, Key=obj.key)
json_data = json_obj["Body"].read().decode('utf-8')
if not json_data:
print("Skipping empty", obj.key)
continue
json_dict = json.loads(json_data)